There is provided an image processing apparatus for performing image processing on image data generated based on an image of a recording medium conveyed in a conveyance direction, comprising: a horizontal edge detection unit configured to detect a horizontal edge included in the image data; a slanted edge detection unit configured to detect a slanted edge adjoining the horizontal edge included in the image data; a calculation unit configured to calculate coordinates of a distal end of the slanted edge, the other distal end of the slanted edge adjoining the horizontal edge; and a coordinate calculating unit configured to calculate coordinates related to an edge of the recording medium based on the horizontal edge and the coordinates of the distal end calculated by the calculation unit.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. An image processing apparatus for performing image processing on image data generated based on an image of a recording medium conveyed in a conveyance direction, comprising: a processor, a circuitry that is configured to perform a method, and a memory that is coupled to the processor and stores a program, that, when executed, cause the processor to alternatively perform the method, said method including the steps of: detecting a horizontal edge included in the image data; detecting a lateral direction edge including the horizontal edge and a slanted edge from the image data; detecting a longitudinal direction edge including a vertical edge and the slanted edge from the image data; detecting a slanted edge that is detected by calculating a logical conjunction of the detection result of the lateral direction edge and the longitudinal direction edge; calculating coordinates of a distal end of the slanted edge, the other distal end of the slanted edge adjoining the horizontal edge; and calculating coordinates related to an edge of the recording medium based on the horizontal edge and the calculated coordinates of the distal end, and correcting a skew of the image of the recording medium based on the coordinates related to the edge of the recording medium, wherein the method is alternatively performed by both of the circuitry and the processor, and wherein the method includes calculating the logical conjunction of the detection result of the lateral direction edge and the detection result of the longitudinal direction edge so as to detect only edges that are present in both the detection result of the lateral direction edge and the detection result of the longitudinal direction edge and eliminate noise that is present in only one of the detection result of the lateral direction edge and the detection result of the longitudinal direction edge.
An image processing system corrects skew in images of recording media (e.g., documents) moving in a specific direction. It detects a horizontal edge in the image, then identifies a slanted edge connected to that horizontal edge using both lateral (horizontal + slanted) and longitudinal (vertical + slanted) edge detection. By logically combining these edge detections (detecting edges present in both detections and removing noise present in only one), the system accurately locates the slanted edge. It calculates the coordinates of the far end of this slanted edge and uses those coordinates, along with the horizontal edge, to determine the recording medium's edge. Finally, it corrects image skew based on the calculated edge coordinates. This processing uses both dedicated circuitry and a processor executing a stored program.
2. The image processing apparatus as claimed in claim 1 , wherein the method includes: detecting the horizontal edge from the image data by a first filter, and detecting the slanted edge from the image data by a second filter.
The image processing apparatus corrects skew in images of recording media. A first filter detects the horizontal edge from the image data, and a second filter detects the slanted edge. The system uses these detected edges to calculate coordinates related to an edge of the recording medium and corrects a skew of the image of the recording medium based on the calculated coordinates. The horizontal edge is detected with one filter, and the slanted edge is detected with a separate, different filter.
3. The image processing apparatus as claimed in claim 2 , wherein the method includes: respectively detecting a plurality of types of the slanted edges by a plurality of filters; performing the detection a plurality of times by using the respective filters included in the second filter.
The image processing apparatus from the previous slanted edge and skew correction description uses multiple filters within the "second filter" mentioned earlier, to detect various types of slanted edges. The system repeats the slanted edge detection process multiple times, each time using a different filter from this collection of filters. This increases the chances of identifying the correct slanted edge, even if it is partially obscured or has variations in appearance. The horizontal edge is detected with one filter, and the slanted edge is detected with a separate, different filter.
4. The image processing apparatus as claimed in claim 3 , wherein the detecting the slanted edges is performed by calculating a logical addition (OR) of detection results of each of the plurality of filters.
In the image processing apparatus that corrects skew, after using multiple filters to detect different kinds of slanted edges (as described above), the system combines the results of each filter. It performs a logical OR operation on the detection results, effectively saying "a slanted edge is present if filter 1 detected it, OR filter 2 detected it, OR filter 3 detected it, etc.". The combined detection result from the multiple slanted edge filters is used, together with the detected horizontal edge, to correct image skew. The horizontal edge is detected with one filter, and the slanted edge is detected with a separate, different filter.
5. The image processing apparatus as claimed in claim 2 , wherein the first filter and the second filter are respectively filters for performing first derivation.
In the image processing apparatus that corrects skew, both the filter used to detect the horizontal edge and the filter used to detect the slanted edge are "first derivation" filters. These filters are designed to identify edges by looking for rapid changes in pixel intensity (e.g., changes in brightness).
6. The image processing apparatus as claimed in claim 1 , wherein the method includes: detecting the horizontal edge by a first filter, detecting a lateral direction edge including the horizontal edge and the slanted edge from the image data; and detecting a longitudinal direction edge including a vertical edge and the slanted edge from the image data; and wherein the method includes calculating a logical conjunction of the detection result of the lateral direction edge by the third filter and the detection result of the longitudinal direction edge by the fourth filter.
An image processing system corrects skew in images of recording media. It uses a first filter to detect a horizontal edge. It then uses a third filter to detect a lateral edge that includes the horizontal edge and a slanted edge, and a fourth filter to detect a longitudinal edge that includes a vertical edge and the slanted edge. A logical conjunction (AND) of the lateral edge detected by the third filter and longitudinal edge detected by the fourth filter is calculated to more accurately identify the slanted edge. It calculates the coordinates of the far end of this slanted edge and uses those coordinates, along with the horizontal edge, to determine the recording medium's edge, finally correcting image skew based on the calculated edge coordinates.
7. The image processing apparatus as claimed in claim 6 , wherein the first filter, the third filter and the fourth filter are respectively filters for performing first derivation.
In the image processing apparatus that corrects skew using separate filters for horizontal, lateral, and longitudinal edges, all three filters (the first filter for horizontal edges, the third filter for lateral edges, and the fourth filter for longitudinal edges) are "first derivation" filters. These filters identify edges by finding areas of rapid change in pixel intensity.
8. The image processing apparatus as claimed in claim 6 , wherein the method includes calculating the logical conjunction of the detection result of the lateral direction edge by the third filter and the detection result of the longitudinal direction edge by the fourth filter and detect an edge that is detected by both of the third and fourth filters.
In the image processing apparatus that corrects skew, after using separate filters to detect lateral and longitudinal edges and performs a logical AND of results: only edges detected by BOTH the third and fourth filters are considered as valid edges.
9. The image processing apparatus as claimed in claim 1 , the method includes: determining whether the edge of the recording medium is chipped based on the detection result of the slanted edge.
The image processing apparatus that corrects skew also checks for damage to the recording medium's edge, specifically for "chipped" edges. It determines whether the edge is chipped based on the detected slanted edge.
10. The image processing apparatus as claimed in claim 1 , wherein a correction process is performed on the image data, in which the image of the image data is rotated about the a point indicated by the calculated coordinates.
The image processing apparatus corrects skew by rotating the image data around a point determined by the calculated coordinates of the detected edges.
11. The image processing apparatus as claimed in claim 1 , wherein the method includes: generating a first formula for expressing the horizontal edge in a two-dimensional plane and a second formula for expressing a straight line orthogonal to the horizontal edge including the distal end in the two-dimensional plane, and finding a point at intersection of the horizontal edge expressed by the first formula and the straight line expressed by the second formula.
The image processing apparatus finds the intersection point of the horizontal edge and a line perpendicular to it. It generates an equation for the detected horizontal edge. Then, it generates an equation for a straight line that is perpendicular to the horizontal edge and passes through the distal end of the slanted edge. Finally, it calculates the point where these two lines intersect.
12. An image processing method of an image processing apparatus for performing image processing on image data generated based on an image of a recording medium conveyed in a conveyance direction, the method comprising: detecting a horizontal edge included in the image data; detecting a lateral direction edge including the horizontal edge and a slanted edge from the image data; detecting a longitudinal direction edge including a vertical edge and the slanted edge from the image data; detecting a slanted edge that is detected by calculating a logical conjunction of the detection result of the lateral direction edge and the longitudinal direction edge, the logical conjunction of the detection result of the lateral direction edge and the detection result of the longitudinal direction edge being calculated by detecting only edges that are present in both the detection result of the lateral direction edge and the detection result of the longitudinal direction edge and eliminating noise that is present in only one of the detection result of the lateral direction edge and the detection result of the longitudinal direction edge; calculating coordinates of a distal end of the slanted edge, the other distal end of the slanted edge adjoining the horizontal edge; calculating coordinates related to an edge of the recording medium based on the horizontal edge and the calculated coordinates of the distal end, and correcting a skew of the image of the recording medium based on the coordinates related to the edge of the recording medium.
An image processing method corrects skew in images of recording media. It detects a horizontal edge in the image, then identifies a slanted edge connected to that horizontal edge using both lateral (horizontal + slanted) and longitudinal (vertical + slanted) edge detection. By logically combining these edge detections (detecting edges present in both detections and removing noise present in only one), the method accurately locates the slanted edge. It calculates the coordinates of the far end of this slanted edge and uses those coordinates, along with the horizontal edge, to determine the recording medium's edge. Finally, it corrects image skew based on the calculated edge coordinates.
13. A non-transitory computer-readable recording medium having stored therein a program for causing a computer to perform a method of image processing on image data generated based on an image of a recording medium conveyed in a conveyance direction, the method comprising: detecting a horizontal edge included in the image data; detecting a lateral direction edge including the horizontal edge and a slanted edge from the image data; detecting a longitudinal direction edge including a vertical edge and the slanted edge from the image data; detecting a slanted edge that is detected by calculating a logical conjunction of the detection result of the lateral direction edge and the longitudinal direction edge, the logical conjunction of the detection result of the lateral direction edge and the detection result of the longitudinal direction edge being calculated by detecting only edges that are present in both the detection result of the lateral direction edge and the detection result of the longitudinal direction edge and eliminating noise that is present in only one of the detection result of the lateral direction edge and the detection result of the longitudinal direction edge; calculating coordinates of a distal end of the slanted edge, the other distal end of the slanted edge adjoining the horizontal edge; calculating coordinates related to an edge of the recording medium based on the horizontal edge and the calculated coordinates of the distal end, and correcting a skew of the image of the recording medium based on the coordinates related to the edge of the recording medium.
A non-transitory computer-readable medium stores instructions for image processing that corrects skew in images of recording media. When executed, the instructions cause the computer to detect a horizontal edge in the image, then identify a slanted edge connected to that horizontal edge using both lateral (horizontal + slanted) and longitudinal (vertical + slanted) edge detection. By logically combining these edge detections (detecting edges present in both detections and removing noise present in only one), the instructions cause the computer to accurately locate the slanted edge. The instructions then cause the computer to calculate the coordinates of the far end of this slanted edge and uses those coordinates, along with the horizontal edge, to determine the recording medium's edge. Finally, the instructions cause the computer to correct image skew based on the calculated edge coordinates.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
January 27, 2016
November 7, 2017
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.